A review of marketing recommendation systems

Abstract

Human nature tends to follow certain patterns in comparing the available options, the recommendation system is one of the secrets of the success of many companies, and this system is a magical marketer for services and products, observes customers and understands their behavior to help them make their decisions. In the era of the Internet and electronic stores, we still need Opinion and advice, the recommendation system that provides us with opinion and advice as e-mails or advertisements has become exclusive messages for us. The recommendation system mainly suggests a list of recommendations in a field, using one of two methods: Collaborative filtering or Content-based filtering or by combining the two methods (hybrid). The research aims to conduct a comprehensive survey of the methods of recommendation systems for the specialized local market, by identifying the most important challenges related to the accuracy of recommendations, problems of cold start, the problem of (NCTR) and data sparsity. By studying many concepts related to the patterns of recommendation systems, and the most important techniques used to measure the accuracy of recommendations and the application of learning methods. Deep Neural Network (DNN) for recommendation segmentation, the main goal of this work is to understand an idea describing niche marketing strategies and challenges. Through the use of various methods and techniques, including the text mining method.